My principal asked the staff to read and respond to “Is Our Grading System Fair” by Monte Syrie. Before reading it, I wrote the following because my principal’s email with this request had the subject line, “Zeroes.” I had thought this all out before but I haven’t shared it very often. I took this opportunity to put it all into words and graphics. After reading the article, I found no reason to change my response to recording or not recording zeroes in the gradebook.
Grades should communicate how much of the assignment a student has mastered. If a student gets two of ten questions correct, telling the student (and parents) he/she got 20% correct is accurate. If a student can answer none of the answers correctly or does not do the work, it is accurate to say we measured the student mastering 0% of the material.
The problem with zeroes is really a problem of labels. Why are most of the letters that label each section of the grading scale scrunched toward the 100? Why aren’t they spread evenly along all the percentages possible?
The rational solution is to move the letters grade labels to match the “Even” section of the following graphic:
However, the common solution teachers are often encouraged to use is to only input grades that fall in the top 50% of the scale. This has the same effect of spreading grades more fairly between A-F, but the common solution doesn’t feel right to a lot of people because then “a student can get a 50% for doing nothing.” Furthermore, this seems to communicate 50% mastery. The CFSID scale makes the situation a bit worse (as seen above) because the letter grades are scrunched even closer to the 100.
If the entire education system is unwilling to re-label the grading scale to make it even, then there’s a fairly easy way to compensate. Grade like you always do, but right before entering grades in the gradebook use this formula on each student’s percentage grade: Take the distance of that grade from 100% (ex: a 70% is 30 away from 100%), and divide that by half (so a 70 becomes an 85%). This makes the scale of possible grades go from 50% to 100%, more closely matching the distribution of letter grades.
And since I’ve already got your attention…
A related problem is the use of categories that divide up the grades in the gradebook. If I understand correctly it was supposed to fix the problem that parents and students were confused by single and double-weight grades. Using categories the way we do now has added to the confusion of students and parents about how much effort to put into each assignment and how much each assignment will affect his/her grade.
It would be better if we used a combination of weighted grades and labeling—as long as each are clearly labeled. It’s a great idea to let students know whether an assignment is major, assessment, or daily, but those should just be informational labels (like hashtags that are so popular today), not separately calculated categories. When we keep major, assessment, and daily grades in separate “buckets,” it becomes nearly impossible for a student to visualize how much effort to put on an assignment to help his/her overall grade in a class. This is because any given assignment’s value compared to an assignment in a different category is never as simple as double or half.
In the following graphic, it takes almost six daily grades in GP1 to equal a major grade, while in GP2 it’s less than three. Even the value of major grades and assessments swap from one grading period. Majors are more important in GP1 while Assessments are in GP2.
By using assignment weights (of 1x, 2x, 4x, etc.) and labeling each assignment as major, assessment, and daily—but keeping grades in all the same “bucket”—it’s easy for a student or parent to see that the work of two daily grades is the same as a major grade (or an assessment) below:
If it’s still too confusing to use single and double (etc.) weight grades, then we could take care to divide assignments into two parts and only ever use single-weight assignments. If we want a test (for example) to be major and double the weight of daily grades, we could have a Part 1 and a Part 2 the students complete and turn in. We would then enter each in the gradebook. (I, however, believe that clear labeling would be sufficient to remove most confusion and would not have the downside of adding even more time spent entering grades in the gradebook.)
One problem this solution doesn’t fix (but also occurs in the current “separate buckets” method) is the relative weight of grades from one grading period to the next. With more or fewer total grades, a daily from one grading period can be worth more or less than from another. This could be solved by using a fixed number of grades required each grading period. Alternatively we could stop splitting grades up by grading period, using semesters or years instead. The downside is that students wouldn’t get as many fresh starts. Those fresh starts, however, are a bit deceiving since grading periods are averaged for the semester or the year anyway.
For all the time and energy I took thinking about these issues and writing solutions, you might think I am passionate about making the changes I proposed. Philosophically, I feel it would be far more beneficial to follow Alfie Kohn’s advice and get rid of grading altogether. I highly recommend reading his articles and/or books to find out why and what educators would do instead.
But if we’re stuck with grading assignments, then it only makes sense to make sure grades do what they’re supposed to: clearly and simply communicate success and learning to students and parents.